Featured Science Paper
On the Robustness of Emergent Constraints Used in Multimodel Climate Change Projections of Arctic Warming
When faced with predictions from a number of different climate models that give a range of estimates of future change, how does one decide which is the most reliable? This is a question that is being encountered more often by climate scientists as the amount of available climate model data increases. It is the main focus of this paper, which is concerned with producing improved predictions of 21st century Arctic climate change. The range in model-simulated predictions is particularly large over the Arctic and reducing this model uncertainty is a major priority due to implications for shipping routes, sea-level rise and access to resources.
It turns out that for some parts of the Arctic quantities such as temperature show clear relationships between biases in simulated present-day climate and predicted future change. For instance, climate models that are colder in their simulation of present-day temperature over the Barents Sea show more future local warming and vice versa. Such relationships are known as emergent constraints, since they emerge from the predictions of multiple climate models and can be used to constrain (i.e. reduce the model uncertainty in) climate change estimates inferred from those models.
In previous work the authors of this paper showed large reductions in model uncertainty in predictions of Arctic warming (up to around a third) by using a simple statistical model fit to emergent constraints in temperature. However, this was based on only one dataset of 24 climate models. With a dataset of this size it is quite likely that spurious emergent constraints could occur by chance. This paper shows that the emergent constraints they analysed previously also occur in a new dataset of 22 climate models, therefore increasing confidence that their approach is robust. The paper is an important contribution to the development of more robust climate predictions for the Inter-governmental Panel on Climate Change.
Bracegirdle, Thomas J., Stephenson, David B.
Journal of Climate, 26, 669–678. 2013